the problem about the version of cuda

[kealennieh]

Thanks for your share. Great job!

I've got one question. Is it really necessary to use tensorflow 1.2 and cuda 8?

I mean the versions of tensorflow and cuda are both higher right now. The combination of my machine is tensorflow 1.13 and cuda 9.1, which can work out in other project. When I run the demo in this project, the result shows "libcudart.so.8.0: cannot open shared object file: No such file or directory". So I guess cuda 8 has been constrained in the program, am I right ?

[garspace]

I countered the same problem. my environment: python3.6 tensorflow1.8 cuda9.0, the error infor:
ImportError: libcudart.so.8.0: cannot open shared object file: No such file or directory
how can i solve it?

[zsh1234567]

I countered the same problem.